Hey, look, there’s a new bicycle signal on Route 9 in Brookline, at the intersection with the Brook House driveway, where they just finished building a new protected bike lane!
But wait – do you see what’s crazy in this picture?
The cars have a green light; the parallel pedestrian crossing has a white Walk signal (in effect, a green light); but the protected bike lane sandwiched between them has a red light!
What’s going on here? If it’s safe for cars to go on the left side of the bike lane and for pedestrians to go on the right side, how can it be unsafe for bikes? Yet, that’s the way this brand new traffic signal operates. Eastbound cars get a green for about 40 seconds, and parallel pedestrians get Walk at the same time, but the bike signal stays red until both are red. Only then – after every direction of traffic has a red light – does the bike light turn green (for 7 seconds, followed by 3 seconds of bike-yellow).
Needless to say, one will be hard pressed to find a cyclist who actually waits for the bike signal to turn green.
Where does this insanity come from?
It starts with a rule – formally, “Interim Guidance” – from an obscure office at the Federal Highway Administration (FWHA) responsible for “Uniform Traffic Control Devices”. That office has decreed that bicycle traffic signals – green, yellow, and red bike silhouettes, which have been safely used for decades in Europe – may be used in the U.S., but with a restriction: they must not be green at the same time any vehicular traffic, including turning traffic, might conflict with bikes. And at this intersection, when Route 9 eastbound has a green light, cars are allowed to turn right into the Brook House driveway, a maneuver that means crossing both the bike and pedestrian crossings. Therefore, per FHWA’s Interim Guidance, the bike signal may not be green while Route 9 eastbound’s vehicle signal is green.
Never mind that in Europe, with decades of experience with bike signals, there is no such restriction. Never mind that the same cars turning right also conflict with the crosswalk, yet it’s considered safe to have a concurrent Walk signal. Never mind that millions of times per day, at hundreds of thousands of intersections across the U.S. where there are no bike signals, bikes follow a green vehicular signal when some vehicles might be turning right. This kind of “permitted turn conflict” is allowed because it is generally safe*. Never mind; a staff person at FHWA has decided that in order to “protect” cyclists, this new device called a “bicycle signal” mustn’t allow bikes to do the same thing they do everywhere else. It must tell the bike riders to wait until all potentially conflicting traffic has been stopped. [*Permitted right turns are generally safe because right turn volume is usually low, and because right turns are made with a tight radius and therefore very slowly. There are some exceptional intersections where right turns pose enough of a danger to bikes that they should be separated from one another in time, but this intersection is not one of those places.]
Has anybody thought about the effect of this rule? Everybody knows that most real-life bike riders will not wait for a red bicycle signal when they can see that parallel pedestrians and cars have a green light and are going. Imagine standing there for 40 s while all the cars on your left go and all the pedestrians on your right go. You’d think, “If I were riding in the street, I’d have a green light and I’d go; if I were riding on the sidewalk, I’d have a walk signal and I’d go. So why should I stand here like an idiot and wait?” It should be obvious that a signal like that won’t actually stop anybody; bikes will go when the cars get a green light. And if it doesn’t change behavior, it won’t affect safety.
So there is no safety effect. But there are two other effects. First, it turns reasonable cyclists into lawbreakers. (I can just hear an official’s response: “Well, if you don’t want to wait for those 40 seconds while your bike signal is red, you could just dismount, step into the adjacent sidewalk, run your bike across the street, and then hop back onto your bike.” Of course.)
Second, it hurts all the other intersection users, adding delay to people in cars, people in buses, and pedestrians. Why? Because complying with this silly rule forces the traffic signal to have a bike-only phase, when everybody else’s signal is red. Without it, the signal cycle would be shorter, and everybody’s delay would be less.
Is there a way around this insane rule? Actually, there are three ways.
Option 1: Get rid of the bicycle light. Or, simpler yet, cover it up. We can hope that within a few years, FHWA will change its guidance and allow the bike signal to be green at the same time as the car signal, and when that happens, we can uncover the bike signal, reprogramming it so that bikes, cars, and pedestrians all get green at the same time. Until then, bikes will just follow the vehicular signals (plainly visible ahead of them), just as they do everywhere else.
Option 2: Petition FHWA for an exception (formally, make a “Request to Experiment”) allowing the bike signal to be green concurrently with the vehicle signal. There’s a good chance such a request would be approved because the right turn is into a driveway (thus, little traffic) and the angle of the turn is more than 90 degrees, forcing drivers to turn very slowly. When Boston built protected bike lanes along Commercial Street in the North End, they petitioned and the exception was granted for three intersections with similar situations, where conflicting turns are into parking lots and driveways.
Option 3: Replace the lenses in the bike signals will simple green, yellow, and red disks, and post a sign “Bike Signal.” FHWA’s rules about bike signals only apply to signals with bike silhouettes; they don’t apply to “normal” signals (disc-shaped green, yellow, red) even if they are signed for bicycle use. That’s the solution used by MassDOT for bicycle crossings in the Casey Arborway project in Boston (see it here in Google Maps).
With any of these options, not only would bikes would be safe and removed from the shadow of illegality, but the car drivers, bus passengers, and pedestrians at this intersection would all benefit from eliminating the 10-second bike-only phase in the signal cycle.
So why didn’t Brookline and its traffic consultants go with one of the three obviously superior options just described? It’s hard to say for sure, but I’ll suggest two likely factors.
First, who can resist the public relations value of including bicycle traffic signals in a project? Nothing says, “We’re progressive and innovative” like bicycle traffic signals.
Second, designers failed to account for expected behavior.
Third, the designers (to their credit) may been looking forward to a future — hopefully not far off — when the FHWA would amend its guidance and allow bike signals with permitted turn conflicts. When that change happens, the Town will want bike signals, but the only chance to get the state to pay for them was through this project.
What should Brookline do now that it’s built? Cover the bike signal heads, or replace the lenses, and retime the traffic signals without a bike phase, with a shorter signal cycle and therefore less delay for everybody. Everyone will benefit, and maybe traffic engineers won’t look so silly.